On July 16, 2026, China’s Cyberspace Administration approved Apple’s generative AI service Apple Intelligence for use on iPhones in the country. Apple will rely on domestic models from Alibaba’s Qwen and Baidu as “technical partners” to power text and image generation features for Chinese users.
This article aggregates reporting from 3 news sources. The TL;DR is AI-generated from original reporting. Race to AGI's analysis provides editorial context on implications for AGI development.
Apple’s long‑awaited AI landing in China is strategically huge: Apple Intelligence will ship, but with its core LLM stack swapped out for Chinese champions Alibaba Qwen and Baidu. That makes Apple the first US platform giant to fully localise its frontier‑adjacent AI features to satisfy China’s licensing regime for generative models, rather than trying to push OpenAI‑ or Google‑powered capabilities through regulatory cracks.
This bifurcation is a milestone in the geopolitical fragmentation of AI. The same marketing label, “Apple Intelligence,” will now refer to fundamentally different systems depending on where the device is sold. For Alibaba and Baidu, being wired directly into iOS, iPadOS, macOS and visionOS is a distribution jackpot, putting their models in front of hundreds of millions of high‑end users and giving them an unprecedented stream of interaction data inside China. For US labs like OpenAI, Google DeepMind and Anthropic, it’s a reminder that the world’s second‑largest consumer market may effectively be closed off, or at least mediated through local partners. Over time, this deepens the split between Western and Chinese AI ecosystems, with implications for benchmark comparability, safety norms and deployment velocity.



